J. Chem. Eng. Data 1986, 3 1 , 86-88
86
UNIFAC with the above interaction parameters. The experimental tie-llne data were also correlated by using the UNIQUAC (3) and NRTL (4) models. The parameters (ul - u]) and (u,,- uu)corresponding to the residual excess Gibbs energy contribution of the UNIQUAC equations and the interaction parameters (gu- gu)and @ - grr)for the NRTL equations were estimated by using the method described by Serrensen et al. (7). The objective function used to estimate the model parameters was
where the indices 2 and 3 represent toluene and isooctane, respectively, and indicates equilibrium values. The S values drop quite rapidly toward a value of 1 at low toluene concentrations. The above results and discussion indicate a low solvent capability of the diethylene glycol methyl ether for the separation of toluene-isooctane mixtures. Acknowledgment
where the summation indices are i = 1, 2 phases, j = 1, 2, ..., N components, and k = 1, 2, ..., M tie lines. Here the x& represent the experimental mole fractions and 2& the calculated ones, and min denotes the minimum value of the summatlon. For the NRTL model, the nonrandomness parameter al was set at a value of 0.2. The UNIQUAC and NRTL parameters, obtained by this procedure, are reported in Table 111, together with the average absolute deviations between the experimental and calculated mole fractions. The experimental measurements on this system showed a small two-phase region at 298 K. According to these results only toluene-isooctane mixtures with less than about 47 mol % of toluene couM be separated by the use of diethylene glycol methyl ether. Figure 3 shows the selectivity of this solvent as a function of the solvent-free mole fraction of toluene in the isooctanarich phase. The solvent selectivity was defined in term of mole fractions as
Sincere thanks to all staff at Kemiteknik for their kind hospitality and assistance and to P. Rasmussen for reading the manuscript. Registry No. DM, 111-77-3; toluene, 108-88-3; isooctane, 540-84-1.
Literature Cited (1) Gani, R.; Brignoie, E. A. Fluidphase Equiiib. 1983, 13, 331. (2) Fredensiund, Aa.; Jones, R. L.; Prausnitr, J. M. AIChEJ. 1975, 27, 1086. (3) Abrams, D. S.; Prausnitz, J. M. AIChE J . 1975, 27, 116. (4) Renon, H.; Prausnitz. J. M. AIChEJ. 1988, 14, 135. (5) Alders, L. "Liquid-LiquM Extraction", 2nd ed.; Elsevier: Amsterdam, 1959. (6) Magnussen, T.; Rasmussen, P.; Fredensiund, Aa. I n d . Eng. Chem. Process Des. D e v . 1981, 20, 331. (7) Sarensen, J. M.; Magnussen, T.; Rasmussen, P.; Fredensiund, Aa. Fluid Phase Equllib . 1979, 3 , 47. Received for review April. 2, 1985. Accepted July 5, 1985. The experimental measurements were performed at Institute for Kemlteknik, Technical University of Denmark, Lyngby, Denmark.
Solutions at Different
Density Study of Temperatures Robert T. Jubin,*t Jack L. Marley,* and Robert M. Councet
Fuel Recycle Division and Analytical Chemistry Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 3 783 1
Demitle8 for the Mg(N0s)2-HNOs-H20system have been measured expermentally; more than 140 measurements were made coverlng the ranges 30-70 wt % Mg(NO,), and 0-40 wt % HNO, at temperatures ol 50-145 'C. A mathematical model for the observed dendty relationships has been developed. Introduction This study was undertaken to provide density data for Mg(N0,)z-HN03-H,0 solutions for the conversion of mass to volumetric flow rates for process simulation and design purposes. This activlty is part of an effort to develope and demonstrate a simple process for the extractive distillation of nitric acid ( 1 , 2). The specific objectives of the study were to obtain density data for solutions containing 30-70 wt % Mg(NO,), and Address for all correspondence: Oak RMge National Laboratory, Post Office Box X, Bulldlng 7601, Mail Stop B, Oak RMge. TN 37831. +FuelRecycb Dhriskin. *Analytical Chemistry Division. Doperatedby Martin Marietta Energy Systems, Inc.. for the US. Department of Energy.
0021-956818611731-0086$01.50/0
0-15 wt % HNO, over the temperature range 50-145 OC. A mathematical model of the density data as a function of the weight percent of Mg(N03), and HN03 and temperature was developed and is presented in this publication. Experhnental Section The experimental procedure is described more completely elsewhere (3). Reagents. Reagent-grade chemicals and demineralized water were used in the preparation of samples and in subsequent analyses. The magnesium nitrate used in these studies was MaHinckrodt Mg(NO3),*6H2O flake. For some tests with high Mg(NO& concentrations, the orlginal hexahydrate was dehydrated to Mg(N03),.(2. 1-2.2)H20. Denslyy Me&w"ents. The dilatometers used in this study were designed to contain -35 mL and were fabricated of Pyrex glass for the study. Calibration was performed by observing and recording the level of the meniscus of water in the graduated regions when equilibrated at 30, 50, 70, and 90 OC. DensHy Measurements, Reagents were weighed directly into 500-mL beakers for the preparation of solutions with magnesium nitrate concentrations of 155 wt % . Accurate 0 1986 American Chemical Society
Journal of Chemlcal and Englnwlng Data, Vol. 3 1, No. 1, 1986 87
Table I. Density of Solutions with 30-35 wt % Mg(NO& *
Table 111. Density of Solutions with 50-55 wt % Mg(NO& wt%
wt%
Mg(N03h 30.86
HN03
30.89
5.15
30.42
10.01
30.46
14.94
29.58
25.79
0.0
temp,O C 50 70 90 50 70 90 50 70 90 50 70 90 50
70 29.98
34.51
33.86
40.02
35.70
0.0
35.43
5.64
33.52
11.64
34.81
16.86
90 50 70 90 70 90 70 90 105 70 90 105 70 90 105 70 90 105
av density, g/cm3 1.252 1.239 1.224 1.286 1.272 1.257 1.318 1.303 1.288 1.353 1.336 1.318 1.414 1.393 1.377 a 1.451 1.430 1.522 1.501 1.284 1.269 1.259 1.322 1.311 1.297 1.344 1.326 1.315 1.396 1.378 1.365
Solution freezes. Table 11. Density of Solutions with 40-45 wt % Mg(N0A
MdNO,), 51.49
HNOn
50.88
4.25
50.85
9.08
50.25
13.94
54.17
0.0
53.13
4.04
54.23
6.02
54.86
9.35
52.24
14.57
55.40
21.06
0.0
temp, O C 90 105 120 90 105 120 90 105 120 ,go 105 120 130 90 105 120 130 90 105 120 130 135 90 105 120 90 105 120 130 90 105 120 130 90 105 120
av density. n/cm3 1.467 1.458 1.449 1.483 1.475 1.461 1.525 1.513 1.499 1.541 1.527 1.510 1.502 1.504 1.495 1.482 1.473 1.514 1.502 1.491 1.479 1.470 1.541 1.529 1.516 1.571 1.558 1.546 1.537 1.566 1.552 1.537 1.526 1.641 1.626 1.616
wt%
MANO,), 40.02
HNOn
40.37
0.0
39.92
4.81
40.09
9.79
39.33
15.46
45.63
0.0
45.10
4.66
44.93
10.03
43.85
15.47
46.21
28.82
0.0
temp, "C 50 70 90 100 50 70 90 100 50 70 90 100 50 70 90 100 60 75 90 100 75 90 105 120 70 90 105 120 90 105 120 90 105 120 90 105 120
av density, g/cm3 1.351 1.340 1.325 1.318 1.359 1.346 1.334 1.329 1.388 1.377 1.360 1.354 1.422 1.414 1.403 1.393 1.454 1.438 1.425 1.411 1.398 1.386 1.376 1.366 1.431 1.415 1.404 1.392 1.456 1.443 1.429 1.481 1.468 1.454 1.585 1.569 1.555
weighing and transfer of reagents were not necessary for this preparation since final concentrations could only be targeted for these reasons: (1) when the magnesium nitrate, nitric acid, and water are being heated and stirred during the dissolution, some acid and water are lost by evaporation, and (2) when transferrlng and heating mixtures containing substoichiometric Mg(N03)2.XH20,(MgX), water is adsorbed while the MgX is exposed to the laboratory atmosphere. The Mg(N03)2*6H20 flake, 70% HN03(sp gr 1.42), and water required to make 190 g of solution were heated and stirred simuttaneously. During the same time, three dilatometers were heated to 90 O C in the oven, allowed to cool for 10 min, weighed, and then returned to the oven. When the salt dissolved and the solution was clear (temperature usually 90-100 "C),the dilatometers were removed from the oven, filled while hot to the graduated region, and weighed. The top closure was then sealed in place. The loaded dilatometers were placed in the oven set at the lower temperature of the range of interest (e.g., the 50 wt % Mg(NO& series, to be heated at 90, 105, 120, and 130 O C , was placed in the oven at 90 "C). After heating at the required temperature for 1.5 h, the liquid level was observed and recorded; then the temperature was raised to the next level, with 1.5 h again allowed for equilibration. At the end of the heating cycle, the dilatometers were removed from the oven and allowed to cool sufficiently so that boiling or gassing would not occur when opened. The hot solutions were immediately transferred to volumetric flasks (200 or 250 mL), allowed to cool to room temperature, and then diluted with water. Aiiquots were analyzed for magnesium by a titration with EDTA; nitric acid was determined by a thermometric titration with sodium hydroxide (4). To prepare solutions having Mg(NO,), concentrations of >55 wt % , use of the substoichiometric MgX was required. Con-
88 Journal of Chemical and Engheerhg De&, Vol. 3 1, No. 1, 1986
Table IV. Density of Solutions with 60-68 wt % MII(NO,)~ w t Yo M g W “ HN03 temp, O C av density, g/cm3 59.82 0.0 90 1.552 105 1.540 120 1.527 135 1.516 145 1.506 59.82 0.0 100 1.558 115 1.546 130 1.534 57.51 4.38 90 1.571 105 1.560 120 1.546 135 1.534 58.90 8.88 90 1.615 105 1.603 120 1.589 130 1.578 61.00 13.84 90 1.668 105 1.656 120 1.641 65.17 0.0 90 1.627 105 1.617 120 1.605 135 1.592 145 1.583 65.97 5.70 90 1.680 105 1.668 120 1.657 135 1.644 65.60 10.44 90 1.700 105 1.693 120 1.676 Table V. Density of Solutions with 70 wt % Mg(NO& w t 70
68.47
HN03 0.0
69.41
4.86
68.79
7.97
Mg(N03h
temp,
OC
90 105 120 135 145 90 105 120 135 120 130
av
density, g/cm3 1.667 1.658 1.646 1.634 1.622 1.703 1.693 1.681 1.669 1.702 1.688
tainers with accurately weighed Mg(NO3),*6H,Q flake were loaded into the vacuum oven set at 41 OC. The trap for removal of water was bmnersed in liqukl nltrop-i, and the system was evacuated. After 5 days of the described treatment, the
oven and contents were brought to atmospheric pressure by backfilling with nitrogen. The contalners were weighed to determine water loss. From the welght loss determination, the apparent formula of the MgX and the percent MS(NO& were calculated. Then the calculated amount of MgX, 70% HNO, (sp gr 1.42), and water were mixed to prepare 200 g of solution. The reagents were dlssokred while heating to 90-100 O C with stirring; then the reagents were transferred to hot dilatometers, equilibrated, and analyzed as described above.
Results The experimental data are grouped according to magnesium nitrate content In Tables I-V. The recommended equation for the density of the Mg(NO3),-HNO3-H2O system is 0 = 0.9280
+ 0.01185M + 0.006215H - 0.0007866T
where D = solution density (g/cm3),M = wt % Mg(N03),, H = wt % HN03, and T = temperature (“C).This equation has a standard devlation of 0.006 125 20 g/cm3 based on 144 experimental points covering the range of 30-70 wt % MdNO,), and 0-40 wt % HNO, over the temperatwe range 50-145 OC. Complete information on the model selection and development is presented in ref 3. R.ghtv NO, Mg(N03)2, 10377-60-3; HNO,, 7697-37-2.
Llterature Clted (1) Counce, R. M.; Oroenler, W. S.;Holland, W. D.;Jubin, R. T.; North, E. D.;Thompson, Jr., L. E.; Hebbie, T. L. “The ExtraDlstlltbn of Nltrlc Acid Using the Two-Pot Concept”, ORNLITM-7982, December 1982. AvaHable from NTIS, the National Technical Informatbn ServIce, United States Department of Commerce, 5285 Port Royal Road, Springfield, VA 22161. (2) Jubln, R. T.; Holland, W. D.;Counce, R. M. “TWOPOT: A Computer Model of the Two-Pot Extractive Dlstlllation Concept for Nitric Acid”, ORNL/TM9241, February 1985. Avallable from NTIS. (3) Jubln. R. T.; Mariey. J. L.; Counce, R. M. “Density Study of the Temary Mixture Mg N0,-H2WN0,”, ORNL/TM-9586, June 1985. Available from NTIS. (4) “Oak Ridge Master Analytical Manual”; Method 1214770, Rev. July 1, 1980; Method 122032, Rev. September 1979, Oak Ridge National Laboratory, Oak RMge, TN. Available from NTIS.
Received for review May 30, 1985. Accepted August 16, 1985. Research sponsored by the Office of Spent Fuel Management and Reprocessing Systerns, US. Department of Energy, under Contract No. DE-AC05-840R21400 wkh Martin Marletta Energy Systems, Inc.
